Mediation analysis in epidemiology: methods, interpretation and bias
Reads0
Chats0
TLDR
The impact of the three main sources of potential bias in the traditional approach to mediation analyses are reviewed and discussed: (i) mediator-outcome confounding; (ii) exposure-mediator interaction and (iii) mediATOR- outcome confounding affected by the exposure.Abstract:
In epidemiological studies it is often necessary to disentangle the pathways that link an exposure to an outcome. Typically the aim is to identify the total effect of the exposure on the outcome, the effect of the exposure that acts through a given set of mediators of interest (indirect effect) and the effect of the exposure unexplained by those same mediators (direct effect). The traditional approach to mediation analysis is based on adjusting for the mediator in standard regression models to estimate the direct effect. However, several methodological papers have shown that under a number of circumstances this traditional approach may produce flawed conclusions. Through a better understanding of the causal structure of the variables involved in the analysis, with a formal definition of direct and indirect effects in a counterfactual framework, alternative analytical methods have been introduced to improve the validity and interpretation of mediation analysis. In this paper, we review and discuss the impact of the three main sources of potential bias in the traditional approach to mediation analyses: (i) mediator-outcome confounding;(ii) exposure-mediator interaction and (iii) mediator-outcome confounding affected by the exposure. We provide examples and discuss the impact these sources have in terms of bias.read more
Citations
More filters
Journal ArticleDOI
Mendelian randomization implies no direct causal association between leukocyte telomere length and amyotrophic lateral sclerosis
Y. Gao,Ting Wang,Xinghao Yu,Raffaele Ferrari,Dena G. Hernandez,Dena G. Hernandez,Mike A. Nalls,Jonathan D. Rohrer,Adaikalavan Ramasamy,Adaikalavan Ramasamy,Adaikalavan Ramasamy,John B.J. Kwok,John B.J. Kwok,Carol Dobson-Stone,Carol Dobson-Stone,William S. Brooks,William S. Brooks,Peter R. Schofield,Glenda M. Halliday,John R. Hodges,Olivier Piguet,Lauren Bartley,Elizabeth Thompson,Elizabeth Thompson,Eric Haan,Eric Haan,Israel Hernandez,Agustín Ruiz,Mercè Boada,Barbara Borroni,Alessandro Padovani,Carlos Cruchaga,Nigel J. Cairns,Luisa Benussi,G. Binetti,Roberta Ghidoni,Gianluigi Forloni,Diego Albani,Daniela Galimberti,Daniela Galimberti,Chiara Fenoglio,Chiara Fenoglio,Maria Serpente,Maria Serpente,Elio Scarpini,Jordi Clarimón,Jordi Clarimón,Alberto Lleó,Rafael Blesa,Maria Landqvist Waldö,Karin Nilsson,Christer Nilsson,Ian R. A. Mackenzie,Ging-Yuek Robin Hsiung,D. M. A. Mann,Jordan Grafman,Christopher Morris,Johannes Attems,Timothy D. Griffiths,Ian G. McKeith,Alan J. Thomas,Pietro Pietrini,Edward D. Huey,Eric M. Wassermann,A. Baborie,Evelyn Jaros,Michael Tierney,Pau Pastor,Cristina Razquin,Sara Ortega-Cubero,Elena Alonso,Robert Perneczky,Janine Diehl-Schmid,Panagiotis Alexopoulos,Alexander Kurz,Innocenzo Rainero,Elisa Rubino,Lorenzo Pinessi,Ekaterina Rogaeva,Peter St George-Hyslop,Peter St George-Hyslop,Giacomina Rossi,F Tagliavini,Giorgio Giaccone,James B. Rowe,James B. Rowe,Johannes C. M. Schlachetzki,James Uphill,John Collinge,Simon Mead,Adrian Danek,Adrian Danek,Vivianna M. Van Deerlin,Murray Grossman,John Q. Trojanowski,J. van der Zee,J. van der Zee,Marc Cruts,Marc Cruts,C. Van Broeckhoven,C. Van Broeckhoven,Stefano F. Cappa,I. Leber,I. Leber,Didier Hannequin,Véronique Golfier,Martine Vercelletto,Alexis Brice,Alexis Brice,Benedetta Nacmias,Sandro Sorbi,Siro Bagnoli,Irene Piaceri,John E. Nielsen,Lena E. Hjermind,Matthias Riemenschneider,Manuel Mayhaus,Bernd Ibach,Gilles Gasparoni,Sabrina Pichler,Wei Gu,Wei Gu,Martin N. Rossor,Nick C. Fox,Jason D. Warren,Maria Grazia Spillantini,Huw R. Morris,Patrizia Rizzu,Peter Heutink,Julie S. Snowden,Sara Rollinson,Anna Richardson,Alexander Gerhard,Amalia C. Bruni,Raffaele Maletta,Francesca Frangipane,Chiara Cupidi,Livia Bernardi,M. Anfossi,Maura Gallo,Maria Elena Conidi,Nicoletta Smirne,Rosa Rademakers,Matt Baker,Dennis W. Dickson,Neil Graff-Radford,R. C. Petersen,D. S. Knopman,Keith A. Josephs,B. F. Boeve,Joseph E. Parisi,William W. Seeley,Bruce L. Miller,A. Karydas,Howie Rosen,J. C. van Swieten,Elise G.P. Dopper,Harro Seelaar,Y.A.L. Pijnenburg,P. Scheltens,Giancarlo Logroscino,Rosa Capozzo,Valeria Novelli,Annibale Alessandro Puca,Massimo Franceschi,Alfredo Postiglione,Graziella Milan,Paolo Sorrentino,Mark Kristiansen,Huei-Hsin Chiang,Caroline Graff,Florence Pasquier,A. Rollin,Vincent Deramecourt,Thibaud Lebouvier,Dimitrios Kapogiannis,Luigi Ferrucci,Stuart Pickering-Brown,AB Singleton,John Hardy,Parastoo Momeni,H. Zhao,Ping Zeng +182 more
TL;DR: Overall, the MR study did not support the direct causal association between LTL and the ALS risk in neither population, but provided suggestive evidence for the mediation role of LDL or TC on the influence of L TL and ALS in the European population.
Book
Etiological Explanations : Illness Causation Theory
TL;DR: Dammann et al. as mentioned in this paper provided a convenient and accessible distillation of the current status of research into this developing field, and added a personal flavor to the discussion by proposing the etiological stance as a comprehensive approach to identify modifiable causes of illness.
Journal ArticleDOI
How does perinatal maternal mental health explain early social inequalities in child behavioural and emotional problems? Findings from the Wirral Child Health and Development Study.
TL;DR: Social disadvantage is associated with higher child externalising behaviour problems score at age 5, and about 40% of this was explained by maternal perinatal mental health.
Journal ArticleDOI
Exposure to Air Pollution during Pre-Hypertension and Subsequent Hypertension, Cardiovascular Disease, and Death: A Trajectory Analysis of the UK Biobank Cohort
TL;DR: In this paper , the authors examined whether ambient air pollution during the prehypertension stage could aggravate the progression from hypertension (HTN) to CVD, and consequent death.
Posted ContentDOI
What explains the effect of education on cardiovascular disease? Applying Mendelian randomization to identify the consequences of education inequality
Alice R Carter,Dipender Gill,Neil M Davies,Amy E Taylor,Taavi Tillmann,Julien Vaucher,Robyn E Wootton,Marcus R. Munafò,Gibran Hemani,Rainer Malik,Sudha Seshadri,Daniel Woo,Stephen Burgess,George Davey Smith,Michael V. Holmes,Ioanna Tzoulaki,Laura D Howe,Abbas Dehghan +17 more
TL;DR: BMI, SBP and smoking mediate a substantial proportion of the protective effect of education on risk of cardiovascular outcomes and intervening on these would lead to reductions in cases of CVD attributable to lower levels of education.
References
More filters
Journal ArticleDOI
The moderator–mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations.
Reuben M. Baron,David A. Kenny +1 more
TL;DR: This article seeks to make theorists and researchers aware of the importance of not using the terms moderator and mediator interchangeably by carefully elaborating the many ways in which moderators and mediators differ, and delineates the conceptual and strategic implications of making use of such distinctions with regard to a wide range of phenomena.
Journal ArticleDOI
Estimating causal effects of treatments in randomized and nonrandomized studies.
TL;DR: A discussion of matching, randomization, random sampling, and other methods of controlling extraneous variation is presented in this paper, where the objective is to specify the benefits of randomization in estimating causal effects of treatments.
Journal ArticleDOI
Marginal Structural Models and Causal Inference in Epidemiology
TL;DR: In this paper, the authors introduce marginal structural models, a new class of causal models that allow for improved adjustment of confounding in observational studies with exposures or treatments that vary over time, when there exist time-dependent confounders that are also affected by previous treatment.
Journal ArticleDOI
Causal diagrams for epidemiologic research.
TL;DR: Causal diagrams can provide a starting point for identifying variables that must be measured and controlled to obtain unconfounded effect estimates and provide a method for critical evaluation of traditional epidemiologic criteria for confounding.
Journal ArticleDOI
A new approach to causal inference in mortality studies with a sustained exposure period—application to control of the healthy worker survivor effect
TL;DR: A graphical approach to the identification and computation of causal parameters in mortality studies with sustained exposure periods is offered and an adverse effect of arsenic exposure on all-cause and lung cancer mortality which standard methods fail to detect is found.
Related Papers (5)
The moderator–mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations.
Reuben M. Baron,David A. Kenny +1 more
Mediation analysis allowing for exposure-mediator interactions and causal interpretation: theoretical assumptions and implementation with SAS and SPSS macros.
Identifiability and exchangeability for direct and indirect effects.
James M. Robins,Sander Greenland +1 more